Acute respiratory failure (ARF) is the leading reason for ICU admission in immunocompromised patients. Usual oxygen therapy involves administering low-to-medium oxygen flows through a nasal cannula or mask [with or without a bag and with or without the Venturi system] to achieve SpO2≥95%.

Oxygen therapy may be combined with non-invasive ventilation [NIV] providing both end-expiratory positive pressure and pressure support. However, in a recent trial by our group, non-invasive ventialtion [NIV] was not superior over oxygen without NIV.

High-flow nasal oxygen [HFNO] therapy is a focus of growing attention as an alternative to standard oxygen therapy. By providing warmed and humidified gas, HFNO allows the delivery of higher flow rates [of up to 60 L/min] via nasal cannula devices, with Fraction of inspired oxygen (FiO2) values of nearly 100%. Physiological benefits of HFNO consist of higher and constant FiO2 values, decreased work of breathing, nasopharyngeal washout leading to improved breathing-effort efficiency, and higher positive airway pressures associated with better lung recruitment.

Clinical consequences of these physiological benefits include alleviation of dyspnoea and discomfort, decreases in tachypnoea and signs of respiratory distress, a diminished need for intubation in patients with severe hypoxemia, and decreased mortality in unselected patients with acute hypoxemic respiratory failure However, although preliminary data establish the feasibility and safety of this technique, HFNO has never been properly evaluated in immunocompromised patients.

Thus, this project aims at demonstrating that HFNO is superior to low/medium-flow (standard) oxygen, minimising day-28 mortality

After discussion at the investigator meeting and based on comments from the Data and Safety Monitoring Board on May 12, 2016, the DSMB has highlighted the need of the interim analysis (already planned) as benefits from high flow oxygen may be observed after 400 inclusions.

Update on June 16, 2017:

The number of patients enrolled is 488 and the inclusion rate is increasing steadily.

The interim analysis has been performed as scheduled and the DSMB decided that nothing should be changed.

Inclusion Criteria:

• Known immunosuppression defined as one or more of the following: (a) immunosuppressive drug or long-term [>3 months] or high-dose [>0.5 mg/kg/day] steroids; (b) solid organ transplantation; (c) solid tumour; (d) haematological malignancy.
• ICU admission for any reason
• Need for oxygen therapy ≥6 Liters/min defined as one or more of the following: (a) respiratory distress with a respiratory rate >30/min; (b) cyanosis; (c) laboured breathing; (d) SpO2<90%; and (e) expected respiratory deterioration during a procedure
• Written informed consent from the patient or proxy (if present) before inclusion or once possible when patient has been included in a context of emergency.

Exclusion Criteria:

• Patient admitted to the ICU for end-of-life care. Do-not-intubate (DNI) patients can be included.
• Refusal of study participation or to pursue the study by the patient
• Hypercapnia with a formal indication for NIV [PaCO2 ≥ 50 mmHg, formal indication for NIV]
• Isolated cardiogenic pulmonary oedema [formal indication for NIV]. Patients with pulmonary oedema associated with another ARF etiology can be included.
• Pregnancy or breastfeeding
• Anatomical factors precluding the use of a nasal cannula
• Absence of coverage by the French statutory healthcare insurance system
• Post surgical setting from D1 to D6

After discussion at the investigator meeting and based on comments from the Data and Safety Monitoring Board on May 12, 2016, as all included patients need to have an acute hypoxemic respiratory failure and at least 6l of oxygen per minute, patients admitted to the ICU to secure any procedure (bronchoscopy etc..) or those not admitted for acute respiratory failure and who undergo intubation, will NOT be included in this trial. Only patients meeting criteria of acute respiratory failure will be included in this trial.